Scraping Banco Mundial.
library(RSelenium)
library(tidyverse)
library(lubridate)
library(rvest)
library(glue)
library(dplyr) #Para calcular la variación mensual
library(jsonlite)
library(plotly)
library(readxl)link = "https://www.worldbank.org/en/research/commodity-markets#1"
eCaps <- list(
chromeOptions =
list(prefs = list("profile.default_content_settings.popups" = 0L,
"download.prompt_for_download" = FALSE,
"directory_upgrade" = TRUE,
"download.default_directory" = getwd() %>% str_replace_all("/", "\\\\"))
)
)
driver <- remoteDriver(remoteServerAddr = "localhost",
browserName = "chrome",
extraCapabilities = eCaps) # Inicia o driver
driver$open() # Abre navegador/browser
driver$navigate(link) # Navega pelo link
driver$maxWindowSize() # Maximiza a janela do navegadorboton_p_mensuales<- driver$findElement(using = 'link text', value = 'Monthly prices')
boton_p_mensuales$clickElement()
Sys.sleep(5)Nos interesan algunos de los datos. No todos.
Importamos el excel descargado.
Eliminamos las primeras 4 filas.
Concatenar las dos primeras filas para que aparezca la unidad de medida.
Modificar el nombre de la primera columna, “Fecha”.
Eliminar todos los M que hacen referencia al mes por “-” para luego agregarle “-01” y así convertirlo a un formato fecha manipulable.
BM <- read_excel("./CMO-Historical-Data-Monthly.xlsx", sheet = "Monthly Prices", skip = 4, col_names = F)
BM_header <- BM[c(1,2), ]
BM_header <- rbind(BM_header, row3 = apply(BM_header, 2, paste0, collapse = " "))
BM_DF <- rbind(BM_header[-c(1:2),], BM[-c(1:3),])
names(BM_DF) <- as.matrix(BM_DF[1, ])
BM_DF <- BM_DF[-1, ]
BM_DF[] <- lapply(BM_DF, function(x) type.convert(as.character(x)))
colnames(BM_DF)[1] <- 'Fecha'
BM_DF$Fecha <- str_replace(BM_DF$Fecha, "M", "-")
BM_DF$Fecha <- lapply(BM_DF$Fecha, function(mes) paste(mes, "01", sep = "-")) %>%
unlist %>%
as.Date()Hay que ver qué se quiere graficar. El gas natural licuado es una buena opción para empezar.
# ggplot(data = BM_DF) +
# geom_boxplot(mapping = aes(x = Mercado, y = Precio, fill = Mercado)) +
# facet_wrap(~ Producto, nrow = 2) +
# theme_light()+
# theme(legend.position="none")
# ## Warning: Can't display both discrete & non-discrete data on same axis## Warning: Can't display both discrete & non-discrete data on same axis## Warning: Can't display both discrete & non-discrete data on same axis## Warning: Can't display both discrete & non-discrete data on same axisCerrar Chrome y eliminar el excel.
unlink("CMO-Historical-Data-Monthly")
driver$close()